Full Carry

A storable commodity can be bought or received earlier and held for later delivery, which requires storage space and ties up capital that has a financing or opportunity cost. Insurance and other applicable expenses can add to the cost of that holding period.

The full-carry comparison puts those costs beside the calendar spread: if the later contract is $5 per unit above the earlier contract and the relevant holding cost is $5, the spread represents full carry in the simplified comparison, provided the dates, unit and commodity quality match. Full carry is more specific than contango.

Contango describes an upward-sloping futures curve, with later delivery prices higher than nearby prices, and the increase can be smaller or larger than the estimated full carrying cost, so an upward curve alone does not establish full carry. Backwardation describes a different price relationship, with nearby prices above later prices, to which short-term scarcity or the value of having inventory available can contribute.

Physical holders can receive a practical benefit from stock on hand that a futures position alone does not supply, and that benefit is often called convenience yield. The NBER discussion of the theory of storage separates interest and storage costs from the value of holding physical inventory, which helps explain why the observed price relationship need not equal simple financing and warehouse charges.

Carrying cost is not constant across time or participants, as interest rates, available warehouse space and insurance terms can change, and one firm's ability to store cheaply does not mean another firm can execute the same transaction at the same cost. Storage feasibility affects the analysis, because a commodity can deteriorate, require special facilities or have delivery-quality rules that limit what stock can satisfy a contract.

A paper calculation assuming effortless storage may omit the most important operational constraint. Transportation and handling can matter when the comparison involves different delivery points or movement between facilities, so those costs need to be identified rather than hidden in an assumed warehouse rate, and contract specifications can make apparently similar commodity quotations non-equivalent.

A spread above estimated carry can suggest a potential cash-and-carry comparison, but it does not guarantee an accessible arbitrage, because execution prices, transaction costs, funding, delivery and operational risks can absorb the apparent difference. Trading two futures months is not identical to owning and storing physical inventory.

A later futures price is also not a guaranteed forecast of the later spot price, since market prices reflect trading conditions and risk as well as carrying relationships. An investor should not interpret the spread as a certain return available merely by waiting.

For a non-finance manager, the concept is useful when comparing inventory timing with forward purchasing, so build an actual holding-cost estimate and match it to the relevant contract interval. That comparison is more informative than assuming a premium for later delivery is unexplained profit or proof that the commodity price must rise.